Rate sheets came in better this morning even though Treasuries did not. The 30-year fixed sits at 6.65%, down 4 basis points from yesterday, the 15-year fixed improved 19 basis points to 5.88%, and the 5/1 ARM gave back almost all of yesterday's pre-Fed spike, falling 41 basis points to 6.58%. Meanwhile the 10-year Treasury pushed up to 4.70%, its highest level of the month. That divergence is not a data error, it is lenders unwinding the defensive margin they built into pricing ahead of the FOMC statement. When a lender does not know whether the committee is about to signal a hike, they pad the sheet. Once the uncertainty clears, the pad comes off, and that is most of what you are seeing in the ARM and 15-year numbers today.
The Fed held the funds rate at 3.50% to 3.75% yesterday for a fifth consecutive meeting, but the vote is the story: 9 to 3, with Cleveland's Beth Hammack, Minneapolis' Neel Kashkari, and Dallas' Lorie Logan all dissenting in favor of a quarter-point hike. Three policymakers dissenting in the same direction has not happened since September 2016. Chair Kevin Warsh declined to call the decision a pause, framing it instead as "a rigorous review of the economic situation." That language matters. A pause implies the next move is a cut; a review implies the next move is genuinely two-sided. Bond markets read it the same way, which is why the 10-year backed up even as the headline decision was exactly what was priced. This morning's June PCE report at 8:30 a.m. Eastern is the immediate follow-up, with consensus looking for headline PCE around 3.7% year over year and core near 3.3% to 3.4%, which would leave the Fed's preferred gauge parked at roughly a three-year high.
For brokers, the practical read is that the fixed-rate side is stuck in a 6.5% to 6.8% band and no single data point is going to break it out this week, but the short end of your pricing menu just got a lot more interesting. With the 5/1 ARM at 6.58% and the 30-year at 6.65%, the spread has compressed to 7 basis points. That is not enough to justify handing a borrower rate risk on a primary residence, and you should say so plainly, because a client who shops elsewhere will hear an ARM pitch this week and needs to know why you passed. On a median-priced $440,600 home with 20% down, a $352,480 loan at 6.65% runs roughly $2,263 a month in principal and interest, about $9 cheaper than yesterday. Nine dollars does not close a deal. What closes deals in this market is structure: a bank statement program that finally documents a self-employed borrower's real income, a DSCR loan that qualifies off the lease instead of the tax return, or an interest-only period that makes an investor's math work at 6.65%.